Statistics from Altmetric.com
Retinopathy of prematurity1 (ROP) has been at the nexus of a progressive understanding of neovascularisation, in large part because of the mouse model of oxygen-induced retinopathy (OIR) developed by Lois Smith.2 This model has been very instructive, because it has crystallised the hypothesis that ROP in general is caused by a two-step or two-phase process, namely oxygen-induced vascular obliteration (phase I) followed by a hypoxia-induced over-production of vasoactive cytokines (phase II), such as vascular endothelial growth factor (VEGF), that is fuelled by increased metabolic demand, decreased oxygen supplementation and widespread local retinal ischaemia created by phase I hyperoxia.
The crucial experiments that truly shaped the concept of phase I and phase II were performed separately in Eli Keshet’s and Lois Smith’s laboratories in 1996, in which they showed that timely injection of VEGF during phase I could prevent retinopathy.3 4 This observation confirms the causative role of hyperoxia in downregulating the tonic production of growth factors critical to retinal development. Hyperoxia creates larger areas of unvascularised retina that are carried into phase II, providing a larger substrate for pathological angiogenesis, as it is the unvascularised retina that secretes excessive VEGF. Additional studies by Chen et al have shown the effect of other growth factors, such as erythropoietin (Epo), in the OIR model.5 Each …